TY - GEN
T1 - Electric Vehicle Charging Robot Charging Port Identification Method Based on Multi-algorithm Fusion
AU - Zhang, Jia
AU - Geng, Tao
AU - Xu, Jun
AU - Li, Yang
AU - Zhang, Chen
N1 - Publisher Copyright:
© 2021, Springer Nature Switzerland AG.
PY - 2021
Y1 - 2021
N2 - Aiming at the problem of plugging and positioning identification in the process of automatic charging of electric vehicles, especially the difficult problem of low efficiency and accuracy of identification in complex operation environment, this article proposes a multi-algorithm fusion of electric vehicle charging port identification method, which can effectively obtain the characteristic information of the round hole of charging port and realize the purpose of automatic identification of charging port by robot. Firstly, an electric vehicle charging port in a complex environment is analyzed and simulated, and camera selection and calibration are explained; on this basis, algorithms based on image smoothing filtering, feature detection segmentation of ROI region, improved Canny edge detection and combined mathematical morphology are proposed to correlate the charging port image respectively, and the features of the target charging port jack are extracted. Finally, the experimental verification of the charging port identification method was conducted for different illumination intensities and different shooting distances. The experimental results show that the identification success rate is 93.3% under the weak illumination and 97.8% under the normal illumination intensity of 4000lx. This shows that the method can effectively improve the robustness and accuracy of the charging port identification.
AB - Aiming at the problem of plugging and positioning identification in the process of automatic charging of electric vehicles, especially the difficult problem of low efficiency and accuracy of identification in complex operation environment, this article proposes a multi-algorithm fusion of electric vehicle charging port identification method, which can effectively obtain the characteristic information of the round hole of charging port and realize the purpose of automatic identification of charging port by robot. Firstly, an electric vehicle charging port in a complex environment is analyzed and simulated, and camera selection and calibration are explained; on this basis, algorithms based on image smoothing filtering, feature detection segmentation of ROI region, improved Canny edge detection and combined mathematical morphology are proposed to correlate the charging port image respectively, and the features of the target charging port jack are extracted. Finally, the experimental verification of the charging port identification method was conducted for different illumination intensities and different shooting distances. The experimental results show that the identification success rate is 93.3% under the weak illumination and 97.8% under the normal illumination intensity of 4000lx. This shows that the method can effectively improve the robustness and accuracy of the charging port identification.
KW - Charging port
KW - Charging robot
KW - Identification algorithm
UR - https://www.scopus.com/pages/publications/85118110542
U2 - 10.1007/978-3-030-89134-3_62
DO - 10.1007/978-3-030-89134-3_62
M3 - 会议稿件
AN - SCOPUS:85118110542
SN - 9783030891336
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 680
EP - 693
BT - Intelligent Robotics and Applications - 14th International Conference, ICIRA 2021, Proceedings
A2 - Liu, Xin-Jun
A2 - Nie, Zhenguo
A2 - Yu, Jingjun
A2 - Xie, Fugui
A2 - Song, Rui
PB - Springer Science and Business Media Deutschland GmbH
T2 - 14th International Conference on Intelligent Robotics and Applications, ICIRA 2021
Y2 - 22 October 2021 through 25 October 2021
ER -